|In Arbitration Between Association of Oregon Corrections Employees and State of Oregon. Ia-13-95.
This matter was heard in Salem, Oregon, on October 5 and 6, 1995. Post-hearing briefs were postmarked to the Arbitrator, following agreed extensions of time, on November 2, 1995, with brief, mutually agreed revisions submitted by each party on November 3. Following the submission of the briefs and initial work on the award, the Arbitrator asked for and received a generous extension of time from the parties in view a family member's involvement in an automobile accident and other personal matters. Due to the length of the award and the number of issues addressed, the Arbitrator has elected to provide a table of contents.
Table of Contents
Background to the Case
I. The New Interest Arbitration Statute
A. The Public Interest and Welfare
1. The Legislative History
2. Plain Meaning
3. Prior Arbitration Authority
4. Other Public Enactments
B. Defining the Secondary Criteria
1. Ability to Pay
2. Retention of Qualified Personnel
3. Overall Compensation
a. "Internal" Comparison
b. Intra-State Comparison
C. Inter-State Comparison
II. The Secondary Issues
A. Incentive Pay
B. Criminal Defense Costs
III. The Wage Dispute
A. Financial Ability
Background to the Case
The Association of Oregon Correctional Employees (AOCE) is an independent union and represents a mixed bargaining unit of security and nonsecurity personnel at the Oregon State Penitentiary (OSP) in Salem, an associated facility in Salem and a very small correctional facility on the Oregon coast, at Tillamook. The State bargains with two other unions representing employees in eight other correctional facilities and a correctional transportation unit. Historically, negotiations between AOCE and the State have been somewhat fractious, resulting in a number of interest arbitration awards, some involving numerous issues.
In 1995 the parties bargained against the background of some dramatic events. A newly-elected, Republican-dominated legislature negotiated with a Democratic Governor to change Oregon public-sector labor law in some significant respects. Ballot Measures affecting public employee pensions and corrections itself passed by popular vote in 1994. An earlier Ballot Measure continued to impact public finances. The legislature sought to limit spending on public employee salaries. The largest unit of organized state employees went on strike near the end of the legislative term. Each of these events is a complicating factor in this case. To their credit, however, the parties resolved most issues in negotiations and bring a small number of questions to this interest arbitration.
The parties proposals will be discussed in detail below. To summarize briefly, the State has proposed a wage freeze for nonsecurity employees, raises (subject to a contingency discussed below) beyond cost-of-living for security employees and an incentive pay proposal with several qualifying criteria. The State' s incentive program would eliminate an existing incentive for the Tactical Emergency Response Team (TERT). The State has also proposed to adjust salaries of nonsecurity employees to account for a potential six-percent reduction that may be caused by a Ballot Measure of dubious constitutionality. The limitation of this adjustment to nonsecurity employees means that, as to security employees, if the Ballot Measure is upheld some or all of the raise otherwise granted will be eliminated. In that event, the State's proposal is, roughly, a wage freeze for all employees in this unit. If, on the other hand, higher courts agree with the lower court decisions that have thus far enjoined the Ballot Measure as unconstitutional, security employee will receive real raises.
The AOCE, briefly, has proposed that the State's salary adjustment mechanism apply to all employees, security or nonsecurity. Thus, this unit would be held harmless from any effects of the Ballot Measure, at least for the term of this contract.(1) In addition, AOCE would give a single cost-of-living increase to the nonsecurity employees. AOCE would give greater-than-cost-of-living increases (which are also greater than the State's offer) to security employees. AOCE proposes an incentive program that does not have as many qualifying criteria as the State proposal. AOCE also would preserve and increase the incentive for TERT team members. Finally, AOCE proposes a system for reimbursing what it views as potentially ruinous costs corrections employees have faced within the criminal justice system, provided those employees are (as a number have been) completely exonerated.
In grievance arbitration the Arbitrator typically sets out the relevant facts and positions of the parties comprehensively and then turns to a discussion of the issues. In this case it is more sensible to present a narrative issue-by-issue discussion, beginning with the parties' differences over the meaning of the 1995 amendments to Oregon public labor relations law.
I. THE NEW INTEREST ARBITRATION STATUTE
In 1995, the Oregon legislature substantially revised its public sector labor relations statutes for the first time since original passage of the law in 1973.(2) Several major changes affect interest arbitration.
At the most basic level, the 1995 legislation changed the structure of the arbitration process to require arbitrators to select the final offer of one of the parties an all points remaining in dispute.(3) The theory behind this design for interest arbitration is that it will cause the parties' proposal to converge, at least on all truly bargainable issues. Thus, the theory goes, use of arbitration will be minimized and collective bargaining will retain pride of place in establishing the parties' relationship. Of course, the validity of this assumption has been questioned.(4) Nonetheless, it appears that in this case the form of arbitration may have contributed to, at least, substantial narrowing of the number of issues submitted to arbitration.(5) The new form of arbitration has also played a part in some of the arguments of the parties, as discussed in more detail below.
The second major change in the interest arbitration statute was a reworking of the criteria for an arbitrator's award. Since the parties offered several significant arguments, and some different approaches, to specific criteria, the Arbitrator will discuss the meaning of these legislative changes point by point.
A. The Public Interest and Welfare
The new interest arbitration statute states that "the interest and welfare of the public" shall be the primary criterium applied by the Arbitrator.(6) The parties do not dispute this, but disagree over the meaning of this phrase.
I. The Legislative History
Both parties have made reference to the legislative history on this subject. The Arbitrator finds this history only slightly helpful. The use of legislative history to illuminate the meaning of the actual words used in legislation has come under attack in recent years from conservative scholars and jurists, notably including Justice Antonin Scalia.(7) While the Arbitrator will not adopt Justice Scalia's extreme position of disregarding virtually all legislative history in almost all circumstances, there is no doubt that some of the criticisms leveled against the formerly routine use of legislative history have merit and that use of such secondary materials should be undertaken with care.
In this case, there are three primary bits of legislative history. First is the actual comparison of the statute as drafted with the terms of the statute as adopted. Second is a critique of the legislation by one of its prominent opponents, Senator Leonard, in the Senate floor debate on the final form of the bill on June 2,1995. Third is a colloquy between two sponsors of the legislation, Senators Derfler and Bryant, in the same Senate floor debate.
The history of the changes in the language of this bill is of some assistance. The parties have both made reference (as do all of the Senators in the June 2 debate) to the changes put in the final bill resulting from Republican sponsors of the bill being compelled to negotiate with the Democratic Governor in order to achieve passage of some public labor law reform. Thus, the changes made reflect the balance struck between a proposal to overhaul the status quo thoroughly and a position more inclined to conserve and maintain existing rules. This provides a helpful context for review of the resulting language, but little more.
Turning to Senator Leonard's critique, the arguments of opponents are one of the least reliable forms of legislative history. Senator Leonard was not engaged in forming the terms of this bill or influencing its nuances when he spoke. He was trying his best to sway votes to kill the legislation or, at least, to make a record usable for future political opposition.(8) Quite simply, his view of the bill was likely to be extreme, since that would make his opposing arguments more appealing. This is not to criticize senator Leonard in the slightest. It is, rather, routine that opponents do not necessarily speak in the most measured and careful terms to the true purposes and form of a controversial bill. Thus, Senator Leonard's remarks do not represent a reliable
construction of this legislation.
In contrast, a colloquy between sponsors was traditionally considered a highly reliable form of legislative history. In this case, however, some caution is in order. First, colloquies have been criticized as an all-too-convenient forum for changing the meaning of a bill in a way that does not reflect the full course of its passage. Here, for example, the colloquy took place on the Senate floor. This exchange of views was neither presented to nor voted upon by the House, nor was it presented to the Governor for signature or veto (nor was it necessarily heard by each Senator).Thus, unlike the words of the law itself, this discussion did not go through the process of bicameral passage and presentation to the executive that creates the law itself.(9)
Second, as already discussed, the bill in this case changed substantially between its initial form and the form being discussed on June 2 following negotiations with the Governor. Here, it is clear that the legislation's sponsors had to compromise. Further, this was a Senate Bill and the effort to secure even more change than was ultimately achieved originated with these Senators. Thus, the danger that these Senators, due to their very status as sponsors of the original legislation, would have a biased view of the result of negotiations to water down that legislation is very real. It is not reassuring, in this context, to have Senator Leonard flatly asserting that the colloquy does not accurately represent the result of the negotiations. While the Arbitrator will not disregard the colloquy between Senator Derfler and Senator Bryant, it will be considered with caution as representing, at best, views which may have informed the vote in the Senate, but not necessarily views of the House or the Governor.
2. Plain Meaning
Turning, then, to the pure language of the bill, the phrase "interest and welfare of the public" is almost wholly uninformative. The problem is not that the concept of "interest and welfare of the public" is unimportant. To the contrary, it is vitally important. It is also extremely general and inherently debateable.(10) Defining the "interest and welfare of the public" is much like defining the term "liberty"; reasonable people are bound to disagree about its meaning. To be sure, one could be faced with a proposal so outrageous that it would be easy to say that it was not in the "interest and welfare of the public." But in most cases, public employee organizations and public managers will each come to interest arbitration with a proposal that is plausible -- after all, both plan to persuade the arbitrator to their view.
Given two proposals that are each within the broad range of what may be called reasonable, it is not clear that the "interest and welfare of the public" will typically yield any basis for picking one proposal over another. Here, for example, both proposals appear to be carefully crafted and in some respects they are very much like each other. No extreme or irrational position has been assumed, by either party, that would make its proposal contrary to the public interest and welfare.
The comments of Senator Bryant in the Senate floor debate are thought to assist in defining this term; but they add only a little. Senator Bryant suggests that the intent of this provision is to cause arbitrators to keep in mind all of the operations of government. In other words, it is not enough, here, to focus on corrections; the award must be consistent with the universe of activities conducted by the State of Oregon. This is helpful in terms of avoiding any excessively narrow misuse of the term "interest and welfare of the public"; but it does not give the term any greater operational significance in deciding between two reasonable, but differing, contract proposals. This consideration is also stated more directly under the ability to pay definition of the new O.R.S. 243.746(4)(b).
3. Prior Arbitration Authority
The Association also argued that the Arbitrator should look to prior arbitral definitions of the public interest and welfare. This is based on a rule of construction that legislators are presumed to know the meaning of established terms of art as construed by adjudicative bodies. The State expresses doubt about the psychological accuracy of this rule of construction; legislators are not necessarily well-versed in the details of decisional law. Of course, this rule of construction is not based on psychological reality. Rather, it is an attempt to impose some discipline on the process of law making by forewarning legislators that they do not write on a blank slate with each new term.
The Arbitrator agrees with the Association that this rule is applicable, but finds the existing arbitral definitions little more precise than the phrase itself. These decisions hold, in general, that quality of public service is in the public interest and that fair compensation is one means of assuring quality of service. Obviously, some balance of fairness to public employees (including a living wage) , quality of service and cost effective service is in "the public interest and welfare." The Arbitrator also agrees with the State, however, that this does not justify throwing money at a problem or assuming that every pay raise results in a marginal benefit in quality of service. Because these observations or "definitions" remain very general, they are not terribly helpful.
4. Other Public Enactments
The State proposes to give the term "interest and welfare of the public" more precise meaning by referring to acts and pronouncements of public bodies and officials. Such behavior is said to necessarily represent the "interest and welfare of the public," as articulated by legitimate governing officials and, therefore, the Arbitrator must defer to their point of view. AOCE's critique of the broadest version of this theory is sound. To put one's ear to the ground for the latest political fad would turn arbitration into something akin to public opinion polling. This would under-mine whatever predictability and stability interest arbitration may otherwise lend to labor relations. It would also mean the "interest and welfare of the public" had no principled meaning, but changed with every shifting political wind.
A more modest version of the State's argument, however, has merit. Political entities have various representative and direct devices for creating binding public policy, or law. Our theory of government is that these devices will, in the end, serve various ends, including the "public interest and welfare."(11)
The Arbitrator's writ is a limited one, focused on resolving a particular labor relations dispute, and no substitute for all other mechanisms of democratic government. Thus, it is not the Arbitrator's province to question, for example, an initiative passed by the voters of Oregon. Nor is it the Arbitrator's right to revise the budget of the State of Oregon or to second guess, for example, final rulings of Oregon courts. Such formal enactments must be presumed to represent the "public interest and welfare." This is not to say that the public employer can simply declare what "the interest and welfare of the public" is and thereby preordain the results of arbitration.(12) After all, interest arbitration is one of the devices Oregon utilizes for creating binding public policy.
In sum, the Arbitrator is required to recognize and accept the legitimacy of public enactments on non-labor relations issues and render an award that is consistent with such expressions of "the public interest and welfare." But the Arbitrator is also charged with preserving the integrity of the interest arbitration process itself; for this process is one more way in which Oregon has chosen to operationalize "the public interest and welfare."
This approach takes us further, but only so far. To the extent the Association has questioned the legislative decision to refund certain moneys to taxpayers, for example, that argument will be given no weight. This non-labor relations decision sets a parameter the Arbitrator must respect. Similarly, the undeniable budgetary impact of Ballot Measure 5 is a "given" which the Arbitrator must accept.(13) But even with such refinements, the "public interest and welfare" does not provide criteria sufficiently specific to resolve this dispute. That Oregon is facing tight financial circumstances, in part due to binding legislative or voter decisions, is important. But it does not automatically mean the lower of any two otherwise similar compensation packages is the only one that can be in the "public interest and welfare." As already discussed that broad phrase encompasses nonfinancial considerations and even within a tight financial circumstance, a choice between two relatively close economic packages might go either way.
Further, AOCE has argued that the passage of Ballot Measures 11 and 17 indicates a strong public interest in supporting a sizable corrections establishment at the State and local level.(14) Simple concepts of supply and demand suggest these measures will put upward pressure on the compensation of Oregon corrections officers. Thus one could conclude from passage of these Ballot Measures that higher pay for corrections officers will more fully serve "the public interest and welfare." The Arbitrator believes this is too facile an approach to the connection between the Ballot Measures and "the public interest and welfare."(15) Among other things, supply and demand relationships in labor markets are notoriously more complex (or "imperfect") than standard macroeconomic theory predicts. Nonetheless, the commitment to a more extensive corrections establishment is another binding public decision which the Arbitrator must somehow respect.
The most confounding measure of the "public interest and welfare" is the passage of Ballot Measure 8.(16) This Ballot Measure has been found unconstitutional by several trial courts under the Obligations of Contracts Clause.(17) These decisions have been appealed and its is expected that the Oregon Supreme Court will be the final arbiter of the fate of Ballot Measure 8. Since the validity of the measure is in serious question it might be thought be that this Act should be disregarded. However, the Arbitrator has no crystal ball revealing the fate of Measure 8. Further, if the measure is upheld both parties have proposed language to "get around" its immediate effect to some degree. This is arguably inconsistent with how the State otherwise approaches "the public interest and welfare" and with the Ballot Measure itself, but this is not a point on which the parties differ in principle. On balance, then, Measure 8 cannot be taken to point favorably to one proposal or the other.
In sum, without consideration of the secondary criteria set forth in the statute, the Arbitrator is still unable to reason directly from the various budgetary decisions and the passage of Ballot Measures 5, 8, 11 and 17, to selection of the offer of one of the parties. Putting all these acts together, one is left with conflicting indications of what will best serve "the public interest and welfare."
Beyond such official actions, the Arbitrator finds little guidance that would make "the public interest and welfare" a precept distinguishing between the two proposals here. Even with the guidance of Ballot Measures and budget enactments (not to mention pending litigation), "the public interest and welfare" is too debateable a concept, standing on its own, to dictate the outcome of this case. The Arbitrator will, therefore, turn to the statute's secondary criteria. As the discussion to this point has indicated, however, "the public interest and welfare," as embedded in binding public enactments, may remain relevant by ruling out certain arguments made by the parties.
B. Defining the Secondary Criteria
As first proposed, the 1995 amendments to PECBA would have made each succeeding criterium in the statute a lower priority. Thus, if "the public interest and welfare" did not resolve a dispute, one would proceed to the second subsection. If this subsection did not resolve the dispute, one would proceed to the third, and so on. See, e.g., Association Exhibit 4, p. , 11. 19-24. This is not the final form of the bill. In its final form, the next six criteria all fall in same category. The statute clearly states that "first priority" is given to the public interest and welfare and "secondary priority" to the other listed criteria (with a residual "other" category given an effective third priority status). The Arbitrator will review these criteria in turn with an eye, first, to resolving the disputes between the parties over how the statutory language is to be interpreted.
1. Ability to Pay
The statutory definition of ability to pay requires "due consideration and weight to the other services, provided by, and other priorities of, the unit of government as determined by the governing body." O.R.S. 243.746(4)(b). As already suggested, the Arbitrator is not entitled to question the budgetary priorities established by the legislature as the governing body of the state. In addition, the statute requires the Arbitrator to respect "[a] reasonable operating reserve against future contingencies." Id. The Arbitrator accepts the State's assertion that a two percent operating reserve is a fair minimum and that an award which will cause or aggravate a lower-reserve is disfavored by this statutory factor. The Arbitrator also accepts the State's assertion that reserves of up to five percent may be considered "reasonable, depending upon all the circumstances.
The Association's arguments in regard to ability to pay (other than suggestions that certain budgetary decision were irresponsible, which the Arbitrator will not entertain) all go to application of this principle and will be considered below.
2. Retention of Qualified Personnel
The next statutory criterium concerns the ability to "attract and retain qualified personnel." O.R.S6 243.746(4)(c) (emphasis added). The parties do not dispute that the State can "attract" qualified entry-level employees in corrections. Thus, the dispute over this criterium focuses on retention.
The parties differ sharply in their approach to analysis of this issue. The State takes a purely retrospective look at recent data on turnover in the Department of Corrections. The Association first takes a longer view of the history, attempting to show that turnover has reached unacceptable levels in the past, presumptively due to wage disparities with County correctional officers. It then attempts to show that recent developments are beginning to create increased turnover and that unacceptable rates of turnover are likely in the future. These argument involve no difference in the construction of the statutory language. Thus, evidence and argument on this point will be reviewed below.
3. Overall Compensation
The statute requires a comparison with the current "overall" or total wage and non-wage compensation of the employees. O.R.S. 243.746 (4) (d) . In itself, this term of the statute gives some weight to the status quo. That is proposals to increase or decrease compensation must find some support in another terms of the statute. This term is also relevant because it defines the basis in existing arrangements for the comparability analysis otherwise required.
The Association has offered a detailed comparison of certain jobs in this bargaining unit with the "overall" compensation of employees of what it considered comparable employers. The State argues that, given the unique posture of this case, overall compensation should not be considered. The State's argument derives from the new requirement that an arbitrator adopt either the State's or the Association's proposal in toto. Again, the purpose of this provision is to cause the parties to narrow (and hopefully eliminate) their differences before submitting a matter to arbitration.
Here, the parties have narrowed their differences, but this means many items of overall compensation are now fixed. In effect, the State is concerned that the Association will prevail in its comparison of overall compensation because it has already agreed to certain levels of non-wage compensation and only its higher wage offer will result in an acceptable comparison of overall compensation.
The Arbitrator appreciates the State's concern. However, there are three major reasons that total compensation must, nonetheless, be compared. First, the very design of the statute creates the tension the State complains about. The statute is intended to encourage settlement of as many issues as possible but also intended to require comparison of total compensation. This combination carries an inherent risk that analysis of the last item or items left outstanding will, with respect to total compensation, be impacted by agreements already made. The statute creates no warrant for the Arbitrator to disregard one command because the parties have complied to a high degree with another part of the statutory design.
Second, assuming the Association pursued a deliberate strategy of taking lower non-wage compensation in hopes of obtaining higher wages, there would be nothing invidious about such an approach. To the extent such choices reflect employee desires there is room for reasonable employees to disagree about what form of compensation is most important. To the extent tactics related to this arbitration could be at work, there is nothing wrong with parties calculating the position they will end up in before the Arbitrator. Put another way, the State has shown no basis for finding a waiver or estoppel in relation to this provision.(18)
Third, it is unclear what limit one could put on a decision to disregard this factor in this case. If negotiation of some items of compensation will always cause those items to be disregarded in total compensation analysis, then the total compensation portion of comparison called for by the statute is effectively repealed, or at least substantially amended, every time parties agree on some but not all items of compensation. Alternatively, parties could refuse to settle on the very basis that this will then undo true comparison of total compensation. This would substantially modify the statutory design of "last, best" offer arbitration. One way or the other, either total compensation analysis or final offer arbitration, or both, will be compromised if the State's position is accepted. The Arbitrator perceives no unique factors in this case that would not be equally applicable in future cases where some compensation items were agreed and others were not.
Thus, the Arbitrator will consider (to the extent data is available) the "overall compensation" of these employees and those who are proper comparisons.
The statute requires comparison with "other employees performing similar services with the same or other employees in comparable communities." O.R.S. § 243.746(4)(e). It then defines comparable communities to be, primarily, those within Oregon of similar population.
a. "Internal" Comparison
The State first asserts that "internal comparisons" -- that is, other employees of the Oregon Department of Corrections should be given greatest weight in this case. The State gives second priority to correction employees of "facility counties"; i.e., counties in which State correctional facilities are located. It gives third priority to a nine county labor market area along the I-5 corridor in northwestern Oregon and final priority to surrounding States, except California.
The Association acknowledges that some internal comparison is legitimate, but disputes the weight it should be given. The Association fell just short of mocking the State's reference to "facility counties," since these include some of the lowest population counties in some remote corners of the State, while virtually all the employees at issue here are in the city of Salem along the urban/suburban I-5 corridor. The Association offers a similar but more limited list of counties on the I-5 corridor than the "nine county" list of the State as its primary set of comparisons and also offers a different perspective on comparison with other states.
For the reasons that follow, the Arbitrator has serious doubts that internal comparisons with Department of Correction employees, as such, should be given an especially high priority in this case. First, the State has otherwise made a persuasive showing that State correction officers tend to be drawn from local or regional labor markets.(19) Oregon (in common with many other states) has a combination of correctional facilities in or near its urban centers and facilities in relatively low-population areas remote from urban areas. Thus, rural eastern Oregon is home to three correctional facilities with about 321 positions. Three counties in this area were also the source of 140 applicants for correctional positions. In contrast, a swath of central Oregon, consisting of twelve counties,(20) holds no correctional jobs and was the source of only a few applicants.(21) Returning to a clear source of corrections applicants, the urbanized, northwestern 1-5 corridor of Oregon accounts for 426 corrections jobs. The number of applicants that should be attributed to this corridor is somewhat difficult to discern from the State's figures.(22) At a minimum, the 259 applicants from Columbia, Multnomah, Clackamas, Marion, Polk, Benton and Linn counties should be attributed to this area. The State's figures separate out another 16 applicants from Tillamook, Yamhill and Washington counties. This is dubious, since this region is only home to seven correctional jobs, at the Pacific coast town of Tillamook. It is probable that applicants from the suburban, Portland-oriented portions of these counties (McMinnville, Beaverton, and so on) account for most of these 16 applicants and were actually for positions on the I-5 corridor. Similarly, the State has given a figure for 44 applicants for a combination of Lane, Douglas, Coos, and Curry counties, when this areas hold 34 jobs at a coastal facility in Coos Bay. Yet it would be surprising if some of these applicants were not persons from the Eugene/Springfield area of Lane County applying for jobs in Salem.(23) Thus, one could find as many as 318 applicants who might have been applying for jobs related to the Portland and Salem facilities.(24) The "true" figure for this I-5 corridor almost certainly lies somewhere between a high of 318 applicants and a low of 259 applicants.(25) In addition, it is not clear from the data whether the small number of jobs at Tillamook and Coos Bay should be regarded as part of a larger "regional" market, or if, instead, distance and transportation factors isolate these facilities from the I-5 corridor to a significant degree.
Put another way, part of the State data analyzes 482 in-State applicants for corrections jobs. Of these, at least 399 (or 83%) were drawn from what the State identifies as local or regional labor markets. At the high end, up to 459 (or 95%) of these in-State applicants could be attributed to local or regional labor markets.(26) That more than 83%, though less than 95%, of this group of applicants appear to be drawn from local or regional markets is a strong argument for looking at State corrections employment in Oregon as involving at least two distinct labor markets: Eastern Oregon, for the facilities at Pendleton, Baker City and Ontario; and some version of the I-5 corridor for the facilities in Portland and in or near Salem (and perhaps the smaller costal facilities). While one could speculate that a significant number of the applicants in eastern Oregon were applying for jobs in the west and visa versa, this appears dubious in light of the large "blank" area with what appears to be an insignificant number of applicants (1.5% of the State total), widely separating the two regions. While the data does not fully reveal to what extent applicants may decide to move from a distance to take a corrections job, it certainly suggests this is a relatively minor phenomena.
Of course, the State's principled reasons for preferring "internal" comparisons go beyond labor market concerns. The State would like to pay employees of its own with similar jobs similar wages. Employees gain transfer rights that may be over-utilized if wide wage discrepancies develop. In collective bargaining, corrections employee in one region are sure to use comparison with another region if that is favorable. Further, the bargaining units are not defined by region and any proposal for a "two-tiered"compensation structure within any a multi-regional unit is certain to draw strong, if not fierce, opposition. All of these concerns are legitimate, though some countervailing issues are obvious. The Association here will argue that its compensation should not be driven by decisions in multi-regional bargaining units that may have been compromised to the detriment of employees in the western labor market. Cost of living may not be comparable by region. Noneconomic preferences for urban, rural or other lifestyles may be significant in employee application and relocation decisions.
Nonetheless, the Arbitrator cannot question the legitimacy of the State's reasons for wanting to give weight to internal comparisons. In particular, the Arbitrator recognizes that the relation of the dynamics of collective bargaining to these apparent regional markets and the three relevant bargaining units is not simple, but complex. Thus, it remains credible for the State to oppose, as a matter of policy and bargaining position, excessive reliance on regional divisions and to insist that internal comparison should be of at least some weight.
There are two difficulties with the State's argument. First, while a "community" is not merely defined by labor markets, the State has not pointed to any non-market factors that would define "communities" in a way that avoids the stark split between Eastern and Western Oregon.
Second, the decisive difficulty with giving pride of place to internal comparisons lies in the statute. However much arbitrators (or the parties) may be concerned with the realities and dynamics of collective bargaining, this interest arbitration is ultimately a creature of the statute. The provision on comparability requires comparison with "other employees performing similar services with the same or other employees *in comparable communities*." O.R.S. § 243.746(4) (e) (emphasis added (**)). Of course, as Arbitrator Lankford has observed, "[o]ther employees of the *same* public employer, who perform the same or very similar services, in the same classification, in the *same* communities . . . actually fit the statutory prescription better than any other sorts of employees.(27) This is entirely sound, but does not address the situation in this case. What if the same employer has employees performing similar or the same services but in communities that are not comparable? The statute, as amended, makes this concern unavoidable when it defines comparable as "*limited to* communities of the same or nearest population range within Oregon." O.R.S § 243.746(4)(e) (emphasis added (**)).(28) Given population discrepancy and distance, it would be, in the Arbitrator's judgment, in direct derogation of this statutory text to find Baker City or Ontario or Pendleton"comparable" to Portland or Salem or the Portland-Salem-Eugene corridor.(29)
It has occurred to the Arbitrator that this statutory definition may have been primarily designed for use with local bodies of government. However, the statute goes on to provide additional guidance with regard to the State as an employer(allowing comparison with other States) and makes no reference to internal comparisons. It follows that the definition of "community" by population must have meaning. Further, that meaning must be more precise than using the full state as a community, when the statute separately provides for state-to-state comparisons. This leaves two ways in which the Arbitrator might, consistently with the statute, look at comparisons with the other Department of Corrections bargaining units. First, the Arbitrator could classify internal comparison as one of the "other," traditional means of interest arbitration analysis under subsection (h) of the statute. This section, however, has been amended so that it can only be used if reference to all the factors in subsections (a) through (g) have not yielded a result. Thus, on this approach, internal comparability of employees performing the same jobs for the same employer would (in this case) be disregarded unless and until the Arbitrator finished the balance of this analysis without reaching a result. Subdividing and prioritizing comparability concerns in this way is not desirable and it is doubtful this was intended by the legislature.
The second approach is to recognize that the other bargaining units have representation in the area otherwise "comparable" to this unit. Thus, the unit involved in this case has 97% of its correctional employees in the Salem area, with the tiny balance in 30 one small costal facility.(30) The OPEU unit has 100% of its employees in the Salem area. Thus, the OPEU unit provides a sound "internal" comparison since it is in the "same" community. In contrast, the AFSCME unit has only 41% of its employees along the I-5 corridor, with 53% being in eastern Oregon and the balance along the coast. Thus, the use of the AFSCME unit for internal comparison is highly doubtful. While it represents a significant number of employees in communities that are the "same or similar," a majority of the employees it represents are from strikingly dissimilar communities (as defined by statute). Thus, the Arbitrator concludes the AFSCME unit does not provide a valid internal comparison, as defined by statute.
b. Intra-State Comparisons
The above discussion of internal comparison calls for even stronger conclusions, with regard to external comparisons. Clearly, the eastern Oregon counties used by the state for comparison are barred, by statute, from that role. The Arbitrator appreciates that the State is concerned that employees enter corrections from these communities and then become part of the overall "system" which it must administer with some sort of internal consistency. The notion that "facility counties" have some significance in this analysis is not, on its face, implausible. The Arbitrator certainly understands the Association's view that county correction officers working the jails in small population, rural counties are inherently not comparable to employees at a maximum security prison, such as the Oregon State Penitentiary in Salem.
The Arbitrator need not review these arguments in detail or weigh them on their "policy" merits; the statute decides this issue in the Association' s favor and does so in unmistakably clear terms. Umatilla, Union, Baker, Grant and Mahleur counties, considered individually or collectively, are not comparable communities, for analysis of the AOCE unit, as defined by statute. They are not of similar population individually to Marion County or collectively to the I-5 corridor. Thus, their consideration is positively forbidden.(31) Both parties have provided slightly different data concerning slightly different sets of counties along the I-5 corridor. These two sets of data concern the only other Oregon public employees the Arbitrator will consider "comparable."
This leaves an issue implied by the State's position on "facility counties." In effect, the State proposed to use the counties in which facilities are located as the primary in-state comparison (after its "internal" comparison). Obviously, the Arbitrator has already ruled this approach out with respect to Umatilla, Baker and Malheur counties. It remains implicit in the State's position that it might view Marion, Multnomah, Tillamook and Coos Counties as "better" bases for comparison than, say, Clackamas County. To the extent the State would continue to take this position, it is insupportable. Again, the State's own evidence points to and documents apparent multi-county labor markets.(32) No matter how one looks at the figures, the overwhelming weight of the corrections jobs in western Oregon lie in or near Salem and in Portland. A very small number of jobs are divided between the two coastal locations. Giving special weight to counties housing facilities with seven (7) or 34 employees, when those counties may be outside or at the very edge of the relevant labor market and do not otherwise appear to be part of the relevant "communities" would, again, be departing from the statutory directive to use population as the characteristic that defines comparability of communities. Further, even without reference to the legislation, it would make no sense at all to give less weight to Clackamas county, which sits astride the interstate system that connects over 400 state correctional employees in relatively close proximity to the north or west (Portland) or south (Salem) , because no facility happens to be located within the Clackamas county borders.
In short, though further data would be helpful to cement this conclusion, it appears that the most relevant county correctional employees are those from the large I-5 counties of Multnomah, Washington, Clackamas, Lane, and Marion. Smaller contiguous counties which may be part of the same "community" or "communities" include Columbia, Yamhill, Polk, Benton and Linn. One or more of these later counties might be excluded on more detailed examination that the record in this case allows.(33) On this record, however, they will be considered as a secondary group of employers relevant to this case, though not as significant as the five urban counties.
Coos and Tillamook counties have not been shown to be part of this labor market and appear geographically separated by coastal mountains and main transportation routes. Absent evidence that these counties are within a market or otherwise within a"community" embracing Salem and Portland (which is to be doubted and was not proved), they are excluded from consideration by statute. Again, viewed as separate communities, these counties are not of the same or nearest population.(34) Thus, for purposes of this award the reliable information put in the record with respect to correctional employees in any of the five urban counties identified above will be considered the primary intrastate basis for comparison, with the immediately adjoining smaller counties, also identified above, used for secondary comparison purposes.
C. Inter-State Comparisons
A related point arises with regard to the states used for comparison. Traditionally, arbitrators have used the four bordering states -- California, Washington, Idaho and Nevada -- as appropriate for comparison for Oregon State employees.(35) In addition, some weight has typically been given California and Washington due to their greater relative population, greater proximity to the employment "centers" of Oregon State government and the similar demographic "weight" of the western and coastal parts of these states. On the other hand, arbitrators have expressed some concern over not giving excessive weight to California due to its size and population and, most recently, the inter-State comparisons have been viewed as secondary to intra-State comparison.
The State argues for a further departure from the weight of prior arbitration awards. It argues that not only should neighboring states be viewed as secondary in importance, but that Nevada and Idaho should be given greater weight in this case (due to the number applicants to the Department of Corrections from each state) and that California should be disregarded as a very different and not comparable state.
The Arbitrator agrees with the State that the recent trend toward viewing inter-state comparisons as secondary is sound. The overwhelming evidence in this case is that the market for these jobs is primarily local or regional. While these local or regional markets cross state boundaries (as with residents of southern Washington applying for jobs in Portland or Idaho residents applying for jobs in Ontario), this is an inter-state aspect of the local or regional character of these markets.(36) Certainly, sufficiently dramatic differences in pay from one state to another may still affect the market. But inter-state comparisons are by their nature secondary.
The Arbitrator also agrees that, for example, Idaho may have greater relevance to employment at the Ontario facility -- which lies on the Idaho border -- than California does. Indeed, it is likely that the Idaho and Nevada applicants have mostly sought employment in eastern Oregon facilities. But this is not persuasive as to what weight should be assigned which state when the unit at issue consists almost entirely of employees in Salem. Thus, the Arbitrator is inclined to utilize the traditional method of looking to all four states as comparable for this unit, with (to put it negatively) lesser weight assigned to Idaho and Nevada. However, the State has also argued that the traditional assignment of greater weight to California is flawed.
Reports on state employee compensation, for example, have argued that Washington and Idaho are the most sound comparison and that California and Nevada are not.(37) Some of the observations made in support of discounting California are provocative.(38) But the Arbitrator hesitates to draw global conclusions about State employees on the facts of this case. This case has demonstrated that corrections employees tend to be drawn from local-to-regional markets. These employees appear to be strongly associated with "communities" that can be otherwise defined on a local-to-regional basis. Further, corrections employment is dispersed around the state in a pattern that is most reasonably viewed as involving different communities, but the same State employer and function. Further, this particular bargaining unit is almost entirely situated within one urban community. That community is proximate to the State of Washington, but not to the States of California,(39) Nevada or Idaho. This combination of factors is almost certainly not characteristic of all other forms of state employment. It may be, taken as a whole, unique. The Arbitrator believes the State's effort to move intra-state analysis forward for all State employees will be rewarded with some measure of success in due time. But the unusual facts of this case caution against making it the flagship for addressing these concerns.
In particular, the Arbitrator believes it is potentially deceptive to look at state correctional employment without considering where within a state correctional employment is located. Thus, for example, Washington is the closest state to the Oregon State Penitentiary. But it is not clear that the Washington state correctional employees represented by the salary figures put in evidence are similarly nearby. If these employees (or a substantial portion of them) are located in Eastern Washington (e.g., Walla Walla), the resulting "average" wages may reflect conditions akin to those in Eastern Oregon, but not at OSP. Similarly, if figures on California correction salaries are driven, in part, the cost of living for correctional employees in the San Francisco bay area (e.g., San Quentin), this would hardly be a valid comparison. In short, while the Arbitrator has reviewed the inter-state figures, these will not be considered significant unless they suggest some serious difficulty with the appropriate intra-state comparison. The Arbitrator emphasizes that this conclusion is driven by the record in this case and factors that may well be unique to corrections.
Thus, by making the inter-state analysis secondary, the importance of deciding how to "weigh" particular states is significantly lessened. First priority, and with it most influence on the ultimate decision, goes to comparisons among employees within Oregon itself. In sum, the Arbitrator will consider the four adjoining states, with lesser weight assigned to Idaho and Nevada, but only to determine if there is some serious difficulty with the conclusions reached based on intra-state comparison.
The State has produced exhibits tending to show that the Consumer Price Index, or CPI, overstates the real effect of rising prices on standard of living. The CPI has been the subject of a growing debate. In some quarters, this is viewed as a hard-minded attempt to refine an inadequate measure. In others, it appears to be an attempt to "soften" the otherwise disturbing and impressive body of evidence that the American standard of living is in the midst of a long period of steady decline. Under the former statute, this Arbitrator would have been required to consider this argument on its merits. That statute referred to "[t]he average consumer prices for goods and services commonly known as the cost of living." This language left room for dispute about what the "real" average might be. The revised statute, however, requires reference to "[t]he CPI-All Cities Index, commonly known as the cost of living." O.R.S. § 243.746(4)(f). For better or worse, this statutory direction is rather precise and will be followed.
To the extent the State is pleading that this element should be given "less" weight because the measure chosen by the legislature is defective, the Arbitrator believes this is a plea for departing from the legislative command. The elements listed in subsections (b) through (g) of the current statute have co-equal status as the list of concerns to be considered after "the public interest and welfare." This does not mean each factor has some mathematically measurable weight in each case. It does mean that an Arbitrator should not consider discounting one factor on the theory that the legislature chose a mistaken criteria or measure. The State's argument, if it truly has merit, should be addressed to the legislature.
The parties have also used two different sub-measures of the CPI All-Cities Index. The Arbitrator has reviewed in detail the figures those measures have produced in recent years and finds the differences to be insignificant. Since the plain language of the statute does not dictate which sub-measure should be used, this factually insignificant difference has no bearing on the proper resolution of this case.
III. THE SECONDARY ISSUES
In requiring the Arbitrator to pick one of the two "packages" on all remaining issues, the Oregon legislature sought to increase the role of the parties in defining their relationship and to decrease the role of arbitrators in that process. This is a valuable and important goal. The benefits of this approach" however, carry a price.
The parties here have presented four basic points of difference. They differ on the wage increases to be given. They differ on how to relate these wage increases to the contingencies created by Measure 8 and the litigation challenging it. They differ on incentive pay. They differ on reimbursement for criminal defense expenses. Clearly, incentive pay and criminal defense expense reimbursement are secondary issues of importance to fewer employees as well as lesser potential fiscal impact on the state. Further, as will become evident, these issues have a weak relationship to the statutory criteria the Arbitrator is directed to utilize. It follows that the difference on wage increases and their relationship to Measure 8 are the key issues in this case. If one position is clearly, superior on these points it may well be more important than the secondary issues, individually or collectively considered.
The parties, however, continue to differ on these issues. They reasonably expect the Arbitrator to address such unresolved points. Further, the full positions of the parties should be analyzed before the final balancing of their merits and demerits.
A. Incentive Pay
The differences in the incentive pay structures proposed by the parties have three aspects. First, the State offers a slightly higher incentive pay package for employees who receive intermediate or advanced BPSST certification than the amounts specified by the Association. Second, the State attaches requirements of seniority, annual uncompensated "education, training and/or community service," and passing a required physical to its incentive pay package. The Association only requires that employees obtain their certificate. Third, the Association maintains and redefines a separate incentive pay category for employees who participate in the TERT team. The State eliminates any incentive pay for this group.
The Association attacks the State's proposal on three fronts. First, it argues that the language requiring "education, training and/or community service," is vague or ill-defined and opens the door to a variety of abuses, disputes and needless administrative headaches. Employees are sure to claim they have done various forms of "education, training and/or community service," which the State may or may not want to recognize for purposes of this program. If the State is restrictive in what it recognizes, disputes are sure to ensue. If the State is not restrictive in what it recognizes, the requirement will not have any rational relationship to the State's interests. Second, the Association views the lack of definition of the physical to be conducted as "critical." In its view, no issue as significant as the form of physical to be administered to employees should be adopted without adequate definition. Third, the Association contends that it is inappropriate to remove the incentive pay for the TERT Team. This group of employees responds to emergency situations in the Penitentiary, exposing its members to increased risks and also imposing significant ongoing training and preparation requirements.
The State defends its suggested requirement by arguing that it is inappropriate to reward employees for the balance of their careers for a one-time achievement in training. Rather, employees should be required to make ongoing efforts at education and preparation if they are to receive premium pay. The State emphasizes that Oregon voters have presented significant challenges to the Department of Corrections. It is appropriate for the Department to expect employees to help it meet these challenges with continuing education, not just a one-time certification.
The Arbitrator agrees with much of the State's argument on this point, but finds the proposed solution is seriously flawed and not justified by the argument. It appears an employee could meet the requirements of this section by serving as an usher in church or a leader in boy scouts (both, presumably, would qualify as "community service") or by taking a class in fly-tying or creative writing ("education").(40) Yet such activities have no apparent nexus with the State's purpose. If it is intended, instead, that a restrictive definition of community service or education is to be applied, it is unclear to the Arbitrator how a line can be drawn without, just as the Association asserts, inviting disputes. In short, if the State offered a more carefully defined program for continuing education or training as a condition of receipt of an incentive pay the Arbitrator would find that reasonable and appropriate. The State's proposal, however, is too general and open-ended to serve its legitimate purpose.(41)
The Arbitrator also agrees that a more specific definition of the physical to be administered is needed. While the Association's fears that this will become a means of terminating employees may be exaggerated, that very fear illustrates how sensitive an issue this can become. Some of the Association's fears may be assuaged by the knowledge that any physical examination would have to be "job related and consistent with business necessity" in order to comply with antidiscrimination law.(42) However, this long-standing requirement has been given a new dimension by the Americans with Disabilities Act. See, e.g. , 42 U.S.C. § 12112(4) (A) and 29 C.F.R. § 1630.14 (c) . Unique aspects of the ADA make it probable physical examinations will have to comply with a new generation of requirements.(43) The Association has good reason to believe more definition of the physical examination requirement is warranted on this grounds alone. Thus, in relation to physical screening as in relation to community service or education, the State's offer lacks adequate definition.
On the other hand, if the Association's proposal truly made incentive pay "permanent" for many employees, it would be objectionable on that basis. However, the ruling in this case only governs the term of the current agreement and will not prevent the State from pursuing its concerns in the next round of negotiation. Thus, the Arbitrator believes the best proposal for the term of this contract is the Association's. The State's legitimate concerns can be grist for the negotiating mill in forming the next contract, but are not well served by the language of the State's current proposal.
While the positions of both parties on the criteria for incentives have some merit, the position of the State on the TERT Team is difficult for the Arbitrator to understand. The Arbitrator took into evidence video tapes of some of the TERT Team's activities within the prison. The Arbitrator agrees with the Association that these tapes vividly demonstrate the increased responsibilities and risks associated with this team.(44) The Arbitrator recognizes that there are non-financial incentives, such as prestige and esprit de corps, for joining the TERT team. Nonetheless, the Association's proposal for some compensation is reasonable and removal of all pay for this activity is not.
The relationship between all these points and the statutory criteria for decision is somewhat attenuated.(45) Certainly, as a form of compensation this item can be used as part of the comparison of "overall compensation" with comparable employers.(46) Most importantly, however, the Association proposal preserves a well-justified item of "compensation presently received by the employees," O.R.S. § 243.746(4)(d); namely, the TERT team incentive. Under this statutory factor, the AOCE proposal is clearly superior.
Thus, the Arbitrator finds the Association has offered the better incentive or premium pay proposal.
B. Criminal Defense
The Association offered testimony concerning several members of the bargaining unit who have been subjected to apparently unjustified criminal prosecutions for conduct in the course of their employment. These prosecutions have resulted in substantial expenditure of personal funds for defense, real personal hardship, and exoneration of the employees. To all appearances these employees engaged in no misconduct of criminal, civil or administrative significance. The Association proposes that employees should be eligible to have their defense costs reimbursed by the State when a prosecution for conduct as employees was not justified. Under this proposal, the State would only be obligated to reimburse employees who were acquitted and were not terminated for misconduct.
The State opposes this provision on various grounds. First, it points out that the Association suggested the State should pay for employees who have experienced this problem in the past. Clearly, it is improper to give this proposal a retroactive effect. Second, it asserts that the problem is not significant because security staff are among the least likely of State employees to face criminal prosecution. Third, the State is concerned that an employee could be acquitted, have a discharge reversed, still be issued substantial discipline short of discharge and be able to claim reimbursement under this provision. Fourth, this provision will give employees strong financial incentives to resist settlement of bona fide charges. Fifth, no other group of state employees has a comparable protection. Sixth, this proposal raises serious policy issues that should only be considered through legislation. Seventh, the State views this as a departure from its proper relation with employees since it is not their "insurer."
This is a difficult issue. The cases presented by the Association involved true hardship visited without warrant on certain employees. This appears to have been caused by an overzealous local prosecutor. The impulse to do "something" for these employees or at least to prevent a repetition of such incidents, is strong. Further, the Arbitrator disagrees with many of the State's arguments. First, while the Association expressed a hope and desire that employees who have been recently prosecuted without justification would be reimbursed, the Arbitrator did not understand this to represent a construction of the contract language. In any event, there is nothing in this language to require retroactive application. It is not for the Arbitrator to say what the State should do as a matter of its sound discretion. But this contract term is viewed by the Arbitrator as prospective only.
Second, it is no answer to employees who have been or may be financially ruined that their misfortune is akin to being struck by lightning. The point of insurance (and the State correctly views this term as creating a form of insurance) is, in part, to spread the risk of low-incidence but highly damaging events. In addition, the Arbitrator is not convinced that these employees face the lowest risk of prosecution; the evidence in the record tends to suggest the opposite. Third, that other State employees do not receive such insurance may be relevant as a comparability argument if its relation to total compensation, on average, were calculated. But no calculation has been offered and this argument does not otherwise undo the Association's demonstration of an unmet need. In fairness to the State, however, the only comparable provision put in evidence was from the State of Idaho. This creates, at best, a weak case of comparability.
Fourth, the Arbitrator does not agree that legislation is required or should be the route to solution of this problem. Employers and employees negotiate various forms of insurance without legislative intervention and there is no reason in principle this could not be such a negotiated item. Finally, the Arbitrator does not agree that there is a "norm" of American employers not insuring their employees for the legal and practical risks of employment. To the contrary, the "norm" is that most insurance (including many forms of assistance that would be social insurance in other industrial nations) are incidents of employment. Medical insurance, workers' compensation, unemployment compensation, disability insurance, pension (or old age insurance), and civil liability insurance are prominent and utterly common examples of how closely we have (for better or worse) connected our insurance system to the employment relationship.
This leaves two arguments made by the State that the Arbitrator believes have some merit. First, the State is correct that this provision would alter the existing "balance" between employees facing criminal and disciplinary charges and the State itself. This might or might not cause increased litigation, but it clearly gives an employee facing charges a stronger bargaining position in relation to all State actors. Of course, there would be nothing wrong with strengthening the bargaining position of innocent employees. But this argument gains force from its relation to the second sound argument made by the State. it is true that employees could engage in substantial misconduct, escape criminal prosecution, and emerge from arbitration with substantial discipline short of discharge. Even more probable, in a close case the risk that an employee might achieve this result, and then be paid defense costs, could be decisive in causing the State to accept a negotiated settlement it found distasteful. In simple terms, while the Association properly wishes to protect employees it views as innocent and wrongfully prosecuted, the State properly fears that well-founded prosecutions of guilty employees (perhaps on difficult evidence) could be adversely affected.
The Association's proposal could be improved (and the State's concern virtually eliminated) by holding that termination or discipline upheld in grievance arbitration would defeat an entitlement to reimbursement. It would be a rare case within these already unusual circumstances, in which an individual deserved to be criminally prosecuted for on-the-job conduct, but did not end up with any employment discipline. Alternatively, some discretion could give appropriate State officials to determine if reimbursement was warranted. Given the proposal as written, however, and the State's sound objections the Arbitrator finds the two positions entirely in balance.
Finally, the Arbitrator finds this proposal has very limited relation to the statutory criteria for decision. Only a limited comparison, to Idaho, was put in the record. No other relationship to subsections (b) through (q) of the statute is evident.
Given equally legitimate needs and objections, as well as lack of a clear connection to the statutory criteria for decision, the Arbitrator find this term does not favor a ruling for either party.(47)
IV. THE WAGE DISPUTE
Both proposals in this case relate their wages to the ongoing dispute over Measure 8. In addition, the State distinguishes security from nonsecurity personnel.
The proposals for non-security employees are easily summarized. The State proposes a wage freeze on the theory that these employees are ahead of appropriate comparable employees. The Association proposes a single cost-of-living adjustment (and thus less than a full COLA for the term of the contract). It characterizes this partial COLA as a "compromise" with its ordinary principles. As to Measure 8, both parties propose salary increases which will hold nonsecurity employees harmless under Measure 8, if it is held valid, but withdraw this wage adjustment if Measure 8 is finally invalidated. Both parties have submitted contract language to allow reopening if the amounts intended as adjustments to deal with Measure 8 are insufficient.
The comparison for security personnel is more complex. The State proposes a five-percent pay increase and one COLA. This is intended to partially "catch up" these employees to what the State recognizes as comparable employers and employees. The Association proposes a greater (but in its view partial) "catch up" through a five-percent increase, a COLA (defined slight differently )(48) and two two-percent increases. The Association's position is driven, in part, by a different set of comparables showing the AOCE security employees further behind.
The Association also proposes that security employees be held harmless from the potential effects of Measure 8. The State, in contrast, has pointedly omitted security employees from the Measure 8 language of its proposal. To be sure, these employee are otherwise receiving raises that nonsecurity employees are not. Perhaps it was thought more necessary to protect employees on a wage freeze from the potential effect of Measure 8. Whatever the motivation for its position, however, the State's position with respect to security personnel becomes internally inconsistent when the variation between validity and invalidity of Measure 8 is taken into account. Simple inspection of the following chart reveals the problem.
COMPARISON OF STATE WAGE PROPOSALS:
Security Personnel and Measure 8
Date of Increase / Measure 8 Valid / Measure 8 Invalid
July 1, 1995 / 5% / 5%
July 1, 1996 / CPI-U (2-4%) / CPI-U (2-4%)
Measure 8 / (-6%) / 0
NET (no compounding / +1% to +3% / +7% to +9%
In effect, if Measure 8 is held invalid the State grants a raise. If Measure 8 is held valid, it extends its wage freeze to the security classifications.(49) In contrast, whether Measure 8 is held valid or not, the Association position remains consistent. The employee raises proposed by the State, though not proposed as a Measure 8 adjustment, are effectively used for that purpose and may be erased by Measure 8. The discrepancy in the State's position is compounded by its effort to offset Measure 8 in other situations. In effect, the State's position turns the security employee's raises into a lottery, which the employees win or lose depending upon the outcome of litigation. This is not based in any principled interpretation of Measure 8, however, since the State is perfectly willing to "offset" that Measure for nonsecurity employees.
This is extremely difficult to understand when one considers that the Measure 8 adjustment is not meant to have any net effect on State spending. Now the State directly pays the employee's share of the pension contribution. Under the Measure 8 adjustment, the State would give employees a salary increase, but then deduct the contribution it is now paying on top of pay, from pay.(50) From the State's point of view this should be roughly budget neutral. That is, the validity or invalidity of Measure 8 should not have a cost impact. Thus, it is difficult to see how a party willing to offer a seven to nine percent pay increase can justify reducing this offer to a zero based on a separate contingency with no net budget impact.
In any event, this is the state of affairs that must be analyzed under the interest arbitration statute's secondary criteria.
A. Financial Ability
The Arbitrator accepts the State's showing that it is at the edge of having acceptable reserves. As a result, this factor is the one most strongly supporting the State. The Arbitrator recognizes that the Association has sincere doubts that every contingent liability identified by the State will come to fruition during this period of time. Predicting the future always leaves some residue of uncertainty. Nonetheless, the State' s showing that its finances are exceptionally tight and its reasonable reserves were under pressure and at risk was persuasive.
While this factor on its face favors the State, the State has created a substantial inconsistency in its position by offering a raise to security employees similar to that requested by AOCE. It is difficult to give a full credence to the State's plea of poverty justifying a wage freeze when it is willing (in one circumstance, but not another) to give security employees up to nine percent in wage increases. In fairness to the State, the Association's last proposal has a greater total cost impact (given the two percent raises for security employees, the COLA for nonsecurity personnel, differences in the incentive pay structure and the like), which must be considered. But the greatest damage the Arbitrator perceives to the State's case is self-inflicted. The stark inconsistency between the two contingencies created by the State's proposal for security employees is extremely difficult to square with any State claim that it lacks funds. If the State can fund the security personnel pay raises if Measure 8 is struck down, then it can at least fund the same pay raises if Measure 8 is sustained.
On balance, the State made a persuasive demonstration that its ability to pay was limited. Then the State agreed to increased pay for security personnel under one circumstance but refused to agree to an identical increase under a financially identical circumstance only distinguished by the legality of Measure 8. This makes it difficult to conclude that the Association's proposal is inconsistent with the State's financial ability. Further, the Arbitrator finds the differences in incentive pay and criminal defense reimbursement were not shown to be financially significant. Thus, the Arbitrator concludes that the limits on the State's ability to pay only clearly favors the State's proposed pay freeze for security personnel. This factor also favors the state, but to a limited degree, with respect to the proposals for security personnel wages.
B. Retention of Personnel
Historically, this bargaining unit has had periods of unacceptable turnover and periods of very low turnover. It appears that the State has recently been successful at creating financial and nonfinancial conditions that reduced turnover and encouraged retention of a veteran employee force. The Association argues that these days of successful retention are coming to an end for several reasons. First, the competition for security employees is destined to increase due to the effects of Measures 11 and 17. Second, the correctional employees in the urbanized counties of northwestern Oregon continue to significantly outpace State corrections employees in total compensation. Third, the most recent evidence is that the rate at which security employees are leaving employment in this bargaining unit is on the rise.
The Association's argument is as good a forecast as any of what might happen in the future. Unfortunately, the ability to forecast accurately is confounded by some of the very factors that concern the Association. For example, Measure 11 and Measure 17 require that "more" be done in corrections. But Measure 5 requires that State and local governments spend less. The State argues that this reflects a public attitude that the State must "do more with less" and that corrections is not immune to this trend. If the State is correct, it is entirely possible that new corrections positions will be created due to Measure 11 and 17, but that money to fund these positions at "competitive" rates will not follow. Thus, ironically, the increased "demand" for corrections activity could actually correspond to downward pressure on non-State corrections salaries. This would probably mean that doing more with less also meant doing it badly. But it is remains unclear whether increasing the volume of corrections business at a time of budget stringency will or will not have the net effect of upward pressure on security personnel salaries.
Second, the State is correct that it has never been a wage leader. Thus, it has shown the ability to retain personnel despite falling consistently behind some of its apparent competitors. On the other hand, one of the contingencies created by the State's offer is an effective wage freeze for security employees. Such a wage freeze would do more than leave the State in its traditional position at the rear of the pack, behind large-county correctional salaries. Rather, this would increase the discrepancy and increase the incentive to consider leaving State employment.(51)
Third, the State is also correct that the evidence of a recent upsurge in turnover is anecdotal and may not be fully reliable. Unfortunately, if turnover is going to increase that would be most likely to be visible after an award favoring the State issued.
Fourth, as to non-security employees the evidence does not show that even a wage freeze would increase turnover. While non-security jobs in corrections can be more difficult and more dangerous than their counterparts in other employment environments, there is no evidence that existing competition is draining employees from the nonsecurity classifications at an unacceptable rate. Rather, as the State suggests, the best evidence on this point suggests that nurses, for example, leave corrections employment primarily because they choose to pursue some other form of nursing. Obviously, if a wage freeze will retain these employees so will a single COLA. Thus, both proposals satisfy this subsection with respect to nonsecurity employees and the State proposal does so more economically.
Once again, the greatest difficulty with the State's position is reconciling its desire and willingness to first give the security classifications an increase that will keep these salaries "competitive" in State's view, but then effectively remove that increase if the irrelevant factor of a valid Measure 8 intervenes.
C. Overall Compensation
Neither proposal makes any change in the status quo with regard to compensation (except for the issue of TERT team incentives already discussed) that does not have a credible justification proffered that is grounded in subsections (b) through (g) of the statute. Thus, "overall compensation" as an independent factor has little significance.
The comparisons that result after the findings already made are, if Measure 8 is held invalid, surprisingly straightforward. The Association has shown and the State does not dispute that the security employees in this unit are significantly behind the comparable corrections employees in the five northwestern, urban counties on the I-5 corridor (Clackamas, Lane, Marion, Multnomah and Washington). Comparing total compensation, the State correctional officers at five years service, for example, are over $1,000 per month below every one of these counties and, on average, over 34% behind.(52) Similar figures attach to other security classifications and years of service.
The second area of plausible comparison involves other counties on the I-5 corridor. Exact figures are not available, but the best proxy for this group is the State's "nine-county labor market" data. This State data here shows a State correctional officer behind comparable employees in all but one of the surveyed counties (Yamhill), and earning 84.9% of the average of these salaries (with insurance and retirement).(53) Sergeants fare worse, earning a lower level than comparable employees in all the surveyed counties and at 80.6% of their average. Id.
The valid internal comparison with the OPEU unit shows that AOCE employees are, at present, behind the OPEU unit. This results, however, from OPEU having a settled contract for the next several years. If, instead, one makes the comparison using the State offer (and assuming Measure 8 does not go into effect), the AOCE security employees would remain very slightly (i.e., less than one percent and with raises at nearly the same dates) behind this comparable unit. Under the Association proposal, these employees would (per the State figures) end up between three- and four- percent ahead of the OPEU unit (with the raises that cause this result staggered between OPEU raises). Though this comparison does not fully discuss overall compensation, it is doubtful including all elements of compensation would dramatically alter this picture.
Thus, one valid, primary comparison strongly supports the AOCE offer. A second valid, more doubt comparison also supports the AOCE offer, but not as dramatically. The third valid, internal comparison supports the State's offer. On balance, the AOCE proposal for security personnel salary increases is, if Measure 8 is invalid, somewhat superior.
If Measure 8 is valid this issue is quite different. Certainly, there is room for doubts about exactly what will happen if measure 8 is upheld. However, it is clear that most collective bargaining in the state is taking place with Measure 8 well in mind and that efforts to offset effective pay increases are certain. To the extent such efforts would be successful with comparable groups of employees, the State refusal to hold security employees harmless from the effect of Measure 8 makes it proposal markedly less comparable than the Association proposal. This State's inadequate and inconsistent response to this risk strongly supports the AOCE proposal.(54)
Finally, comparison with the four bordering states shows that Oregon is significantly behind California in total compensation, very similar to Idaho and Nevada and noticeably ahead of Washington. Without engaging in the exercise of trying to "weigh" these figures, it seems clear that they are consistent with either maintaining the status quo or with increasing compensation modestly. That is, given the many uncertainties attached to this data, it does not argue strongly against the conclusion drawn from the interstate comparisons.
As to nonsecurity employees the State has offered data that suggests these employees are paid more than comparables at, for example, the OPEU unit.(55) This evidence is not quite as clear with regard to employees in the five-county correctional systems. For example, AOCE nurses have a lower wage minimum than employees in all five counties and a higher maximum than employees in Washington county only.(56) However, when insurance and retirement is added in, the AOCE nurses are better paid than corrections nurses are in three of the five counties at range minimum and two out of five at range maximum. Id. On the other hand, AOCE corrections counselor are paid higher than comparable employees in each county, in salary alone and salary plus insurance and retirement, at both range maximum and minimum. Id.
The Association's response to this is twofold. First, the Association points out that it is only requesting one cost-of-living increase for these employees. Further, it analyzes what it considers comparable employees in the five urban, I-5 counties to show that they are receiving yearly increases that appear to be COLAS.(57) Thus, comparable employees who may be paid less will be "catching up," with AOCE employees over the term of this contract. Second, the Association emphasizes that these employees are not truly comparable to some of the State's comparisons because they practice their profession or trade in the special, difficult and sometimes risky circumstances of a maximum security prison. In support of this point, the Association offered vivid testimony concerning the responsibilities placed upon and risks faced by correctional nurses, in particular. The Arbitrator agrees that corrections nurses perform in an environment that calls for different compensation than many other nurses. The same may be true of many of the other classifications involved in the nonsecurity branch of this case.
Tellingly, however, both arguments by the Association as much as admit the basic point made by the State. Further, the State's evidence comparing these employees to the OPEU-represented employees at OSCI and the correctional employees in the five urban counties, are with fully comparable groups of employees. Corrections nurses are compared to corrections nurses, and so on. It is more helpful that the Association did not ask for a full set of cost-of-living raises. However, this concession does not appear proportionate to the extent of the gap in many of these cases. To give just two examples, AOCE corrections counselors are paid (in wages, insurance and retirement) roughly six percent more than the highest comparable employees in county corrections (Lane County).(58) In relation to Multnomah County, the gap is 18%. Id. On the other hand, the gap in relation to such employees at OSCI is only about three percent and this appears to be roughly the same for many of 20 comparable positions.(59) As to this "internal" comparison the Association's offer might eliminate the gap and would probably reduce it significantly.
On balance the Arbitrator concludes the State has shown these employees to be paid, in comparison to the most relevant State and county employees, generally above the appropriate averages. Further, the Association's offer does not predictably eliminate this gap. In sum, while the issue is close the Arbitrator finds the State's position the more reasonable of the two in comparability for nonsecurity employees.
Obviously, the CPI is included in the statutory factors on the assumption that unless justified by other considerations, wages should not be eroded by inflation. By the same token, the assumption is that special justification must exist for exceeding the CPI. Here, as to security personnel the State effectively admits a raise above CPI is warranted. However, the State then withdraws that raise under circumstances with no justification in the statutory criteria (i.e. , the validity of Measure 8) . The Association effectively concedes a raise lower than that indicated by the CPI is warranted as to non-security employees, but then offers to limit raises in a way that does not quite address this issue. In short, CPI, as a factor, points in the same directions as comparability.
In sum, the Arbitrator generally finds that AOCE proposal superior with regard to comparability of overall compensation for security employees and the State proposal superior with regard to nonsecurity employees. The State's advantage with regard to nonsecurity employees is close (which is hardly surprising when a wage freeze is being compared to one cost-of-living adjustment in an age of low inflation). The AOCE's advantage with regard to security employees is stronger, especially if Ballot Measure 8 is found valid. Again, the State proposal is deeply flawed in its treatment of security employees if Measure 8 is held valid. Further, security employees are predominant in this unit. Thus, the statutory factor of comparability, on the whole, strongly favors the AOCE proposal. The Arbitrator finds that the factor of ability to pay favors the State. However, given the State's willingness to fund most of the pay increase for security employees proposed by AOCE (if Ballot Measure 8 is held invalid), this is not as weighty a factor as comparability. Neither retention of employees, nor overall compensation as an independent factor, nor the CPI strongly favors either party's wage proposals. Finally, the AOCE has the superior proposal on the issue of incentive pay. On balance, the Arbitrator concludes that the AOCE offer must be accepted.
The Arbitrator selects and approves and directs the parties to implement the "last, best" offer of the Association.
William P. Bethke
February 5, 1996
For AOCE: John Hoag, Esq., Hoag Garrettson Goldberg & Fenrich
For the State of Oregon: Gary M. Cordy, Esq., Assistant Attorney General
1. This assumes that if the Ballot Measure is lawful, the salary adjustment is lawful. As discussed below, this assumption is not certain, but is utilized by both parties. Thus, whatever doubts there may be about this approach, it forms no basis for the Arbitrator favoring one proposal over the other.
2. For a perspective on the original legislation see Widenor, Public Sector bargaining in Oregon: The Enactment of the PECBA (LERC Monograph Series 1989).
3. "[T]he arbitrator shall select only one of the last best offer packages submitted by the parties . . . O.R.S. 243.746(5).
4. See, e.g., Coleman, Jennings & McLaughlin, Convergence or Divergence in Final-Offer Arbitration in Professional Baseball, 32 Indus. Rel. 238 (1993).
5. On the other hand, there is some suggestion that these parties are entering a more "mature" phase of their collective relation and much of the narrowing of issues may be attributable to that development.
6. "Arbitrators shall base their findings and opinions on these criteria giving first priority to paragraph (a) of this subsection and secondary priority to subsections (b) to (h) . . . as follows: (a) The interest and welfare of the public." O.R.S. 243.746(4).
7. For a sample of the debate see, e.g., Wald, The Uses of Legislative History in Statutory Interpretation Cases in the 1992 U.S. Supreme Court Term, 23 S.W.U.L.Rev. 47 (1993); Breyer, On the Uses of Legislative History in Interpreting Statutes, 65 S.Cal.L.Rev. 845 (1992); Mikva, A Reply to Judge Starr's Observations, 1987 Duke L.J. 380.
8. See, e.g., Mikva, supra. Judge and former Representative Mikva gives a rueful account of the judicial uses of his opposition to the passage of RICO.
9. Compare INS v. Chadha. 462 U.S. 919 (1983) (finding federal "legislative veto" statutes unconstitutional for lack of bicameralism and presentation).
10. The political theorist William Connolly, building on the work of British philosopher W.B. Gallie, has argued that many of the most important terms in public debate are "essentially contested." That is, while these concepts are widely shared they are also complex; involve both descriptive and normative dimensions; can be shaped in subtlety different fashions to meet different states of "evidence"; and can even influence what counts as evidence. Connolly, The Terms of Political Discourse (1974).
11. Thus, the preamble to the United States Constitution identifies, among other goals, "promot[ing] the general Welfare."
12. The original bill may have been intended to do just that. It required arbitrators to follow "[t]he interest and welfare of the public as determined by the governing body of the public employer . . ." Association Exhibit 5, p. , 1. 23.
13. Ballot measure 5 is a tax limitation measure that places ongoing pressure on State and local government budgets.
14. These Ballot Measures increase terms of imprisonment predictably requiring more prison space and more correctional employees, and also require prisoners to work. This requirement will predictably increase correctional expenses.
15. It is fair, however, to argue from the predictable effects of the Ballot Measures to the conclusion that current figures on, for example, employee turnover, may understate the real situation during the life of this contract. Again, this may be helpful, but relates to one of the secondary criteria established by statute.
16. Ballot Measure 8 attempts to outlaw the common practice of negotiating for public employers to "pick up" what would otherwise be the employee contribution to the state pension system.
17. U.S. Const., Art. I, Sec. 10.
18. This highlights, however, one reason last, best offer arbitration may not completely fulfill its theoretical promise of eliminating all genuinely negotiable items. Parties are still likely to calculate how they can appear less unreasonable than their counterpart while not compromising as fully as they might if faced with the alternative of a strike or lockout. See n. 4.
19. In the State's words, there is a "direct correlation between the sites of state facilities and the number of applicants *from the region surrounding the facility*." Employer's Post-Hearing Brief at 42 11. 18-19 (emphasis supplied (**)).
20. Hood River, Wasco, Sherman, Gilliam, Morrow, Wheeler, Jefferson, Crook, Deschutes, Klamath, Lake and Harney.
21. The data in Exhibit E-e6 contains different groupings of counties in different portions. As a result, it is difficult to tell how many or if any applicants came from certain counties that are grouped one way in one portion of the exhibit but may be grouped differently or omitted entirely in another portion. The 12 county area was clearly the source of seven (7) applicants and perhaps no more.
22. All figures in this paragraph are derived from Employer Exhibit E-e6.
23. The State also gives a separate figure of 16 applicants from Josephine and Jackson Counties, along I-5 at the southern margin of the State. As all the available corrections jobs are somewhat remote to these counties, the Arbitrator can draw no firm conclusion regarding where these employees were looking for work.
24. One could also assimilate the 41 jobs at Tillamook and Coos Bay to this figure. That would have little if any effect on analysis and may not be justified. It is not clear from the data provided or from "common sense" whether individuals readily leave the coast for urban jobs and visa versa, or not.
25. At least some of the applicants were probably seeking jobs in the coastal facilities. As already noted, the State data does not subdivide these in a way that shows how localized or integrated these jobs are with the "I-5" labor market.
26. In theory the percentage could be higher, if one assumes a sharp separation between applicants to the Coos Bay and Tillamook facilities and applicants to the Salem-Portland facilities. There is no data in the record to support this, however, and it is at least equally possible the percentage could be driven below 95% by a variety of factors.
27. OPEU, SEIU Local 503 and State of Oregon (Lankford January 26, 1994) 21 (emphasis in original (**)).
28. This definition gives the statute far more precision than its predecessor. Thus, Professor Snow's accurate description of the open-endedness of pure "comparability" analysis is no longer a fair description of the statutory scheme. See OPEU and State of Oregon (August 7, 1995), especially at 22-27.
29. Distance is relevant to deciding what counts as a "community." Thus, the statute continues to offer some flexibility with regard to comparables. Here, the evidence supports a regional definition of the "community" relevant to this case. It would be torturing the statutory text, however, to find all of Oregon a single "community" for purposes of corrections employment. Once this extreme construction is rejected, dividing Eastern Oregon from the West is inevitable.
30. These figures are derived from the summary of employment in the 1993-95 biennium in Employer's Exhibit E-e6.
31. Neither party offered comprehensive population data for Oregon counties. However, the limited data in the record is sufficient to support this conclusion and the Arbitrator believes any further data could only strengthen the basis for this conclusion. Indeed, the State's argument may be more troubled than the discussion in the text indicates. County boundaries tend to retain their original demographic significance in rural areas. Thus, the "community" in such cases may well be the county. In contrast, urbanized areas often blur or cross over these traditional boundaries. The "community" in an urban area will often plausibly cover a number of cities and counties. The proper analysis under the statute, therefore, could arguably look to the population of the Salem-to-Portland (or even Eugene-to-Portland) region as one community, but to Baker, Umatilla and Malheur counties as three separate communities. Obviously, under this approach the population disparities are even more overwhelming.
32. The Arbitrator has some doubts about the State's choices among possible groupings of counties for purposes of these comparisons. These doubts, however, run in the direction of keeping suburban counties attached to their urban areas and not pairing them with adjoining coastal counties. Also, the northwest I-5 region may be properly defined as more than one "community" for a variety of reasons. However, in this case any subdivision is likely to result in separate communities that remain comparable, in population, to each other. Thus, this concern does not directly affect consideration of, at least, the five largest urban counties. Even if it were proper to subdivide "communities" to the county level, these counties would remain comparable. At this level, however, some doubt would exist as to the smaller "I-5" counties.
33. If, for example, it was clear that employees were not applying to the state from one of these smaller counties that would suggest they were not legitimately part of the same "community." similarly, non-market factors, including characteristics of the corrections establishment in smaller counties, might justify treating these as separate "communities." If these are separate communities, then they may be ruled out by statute due to lower population.
34. One could use the employees at Coos Bay to compare with this unit's employees at Tillamook. However, the Arbitrator does not understand the State to assert that the situation of these seven (7) employees has significant weight in this case.
35. It does not appear that parties have argued to arbitrators that other states should be considered. However, the statute does not impose this limitation and Professor Snow's reminder that a broader set of comparisons may be appropriate remains applicable to considering what states are truly comparable to Oregon.
36. The most sensible approach would be to use the relevant local or regional areas within, for example, Washington, for comparison on terms similar to the comparisons with country correctional employees in Oregon. The statute does not countenance this approach.
37. See, e.g., exhibit E-e2 at pages 17-20.
38. But see Association Exhibit 16 (comparing border populations, border traffic, inter-state visitors, miles of shoreline, ports, and, most entertainingly if not most persuasively, "stuff").
39. However, the one group of intrastate applicants that the Arbitrator has not discussed in detail to this point are the 16 applicants from southeastern Oregon, on the California border, but also astride I-5.
40. Contrast, e.g., Association Exhibit 30 at 1-2 (Clackamas County Negotiated Incentive Program).
41. The Association argues that the State engaged in bad faith bargaining by inserting some of these conditions in its last offer without substantive discussion of a continuing education program at the bargaining table. The State asserts that it is in the nature of "last, best offer" arbitration that the very last offer may have some material difference from what has gone before. Since the Arbitrator agrees with the Association that these proposals are deficient, it is unnecessary to rule on this issue. However, it appears the Association may be open to discussion of more carefully defined conditions for incentives and, clearly, more extensive discussion in bargaining may help take the rough edges off the State's proposal.
42. Traditionally, "discriminatory impact" cases alleging sex discrimination have challenged and helped define physical requirements for public safety employment. Presumably, the State would design its physical to avoid this well-known pitfall.
43. Thus, examinations must take into account possible reasonable accommodations of employees with disabilities. See, e.g., 29 C.F.R. § 1630.14(c)(1)(i). The issue of reasonable accommodation necessarily implies some attention to what would constitute, under various circumstances, an "undue burden," as well as some definition of the "essential functions" of the job and some process of consultation with employees regarding these issues. See, e.g., 42 U.S.C. §§ 12111(8), (9), and (10). In addition, the ADA recognizes, as a matter of antidiscrimination law, a protected privacy interest in the contents of certain examinations. See, e. g. , 42 U.S.C. § 12112 (c) (3) and 29 C.F.R. § 1630.14 (c) (1) . This requirement arises in part from the belief that information regarding disabilities is often misconstrued or misused, causing resulting discrimination. These are fairly obvious requirements of the ADA. Others may have escaped the Arbitrator's attention.
44. The tapes were returned to OSP after the Arbitrator viewed them because one of the tapes was considered relevant to investigation of an inmate accused of a murder which took place within the Penitentiary after the arbitration hearing.
45. Any of these points could be treated as covered under subsection (h) of the statute. In that case, however, they would only have relevance if a decision could not otherwise be reached.
46. A number of Counties offer correctional security employees what the Association terms "automatic incentives." See Association Exhibit 22 (detailing "Automatic Incentives" for employees at various levels of seniority in Clackamas, Lane and Marion Counties).
47. In the absence of a clear relationship to subsections (b) through (g) of the statute, this term could be related to either subsections (a) or (h). Under subsection (a) a plainly inappropriate proposal could be said to adversely impact "the public interest and welfare," giving greatly increased weight to this otherwise secondary factor. On the other hand, unjustified State resistance to a sound proposal would at least be one of those "other factors" to be considered if an only if no decision was dictated by subsections (b) through (g). Given that both the proposal and the resistance to it have merit, there is no need to decide which statutory pigeonhole it might otherwise fit.
48. The State uses the CPI-U measure while the Association uses CPI-W. The Arbitrator has inspected the data on recent changes in these measures and finds the differences insignificant. The State proposes to cap the COLA at four-percent. The Association caps the COLA at five-percent. Given the near-term prospects for inflation the Arbitrator finds this difference unlikely to have any real effect.
49. Given the various dates involved and interest, a valid Measure 8 might well eliminate the one-to-three-percent shown in the chart.
50. This assumes such a Measure 8 "offset" is constitutional if Measure a is found valid. That assumption is grounded mostly in the reality that both parties here have made this proposal and the Arbitrator is directed by law to pick one of the two offers. The Arbitrator makes no finding regarding and does not believe it prudent to offer an opinion on this issue of law at this time. The Arbitrator notes, however, that Contract Clause cases often turn on the difference between prospective and retrospective effects. Measure 8's attempt to forbid offsetting pay increases is, among other things, more retrospective in nature when applied to a bargain settled before the Measure comes into effect.
51. The evidence shows that ongoing or recently completed negotiations in the five most comparable counties have included a wage increase.
52. See Association Substitute Exhibit 22.
53. Exhibit E-e2O. The State survey does not show overall compensation within the meaning of the statute. The Association exhibit suggests that the State consistently fares worse in a complete overall compensation comparison than it does in a salary comparison. See Association Substitute Exhibit 22 (comparing "ER Disposable Income" with "ER Cost for Total Compensation").
54. There may be a tendency for the both parties to discount this risk since district judges have ruled Measure 8 unconstitutional. Without implying any judgment on the merits of this dispute, it is the Arbitrator's experience that confident predictions of the behavior of courts of last resort in cases of first impression are foolhardy.
55. See e.g., Exhibit E-e9 (in 19 out of 20 non-security classifications, OSP employees are paid more than employees at the OPEU corrections unit).
56. Exhibit E-e19.
57. Association Exhibit 28. This exhibit was said to represent "courthouse," non-security employees at the county level.
58. See Exhibit E-e19.
59. Exhibit E-e9.